Re: Proglem doing CA calls after catching a signal

[Eric Norum <norume@aps.anl.gov>]

Chris Larrieu wrote:
> Isn't the crux of the problem that there's no way to specify/predict 
> which thread should handle a signal?

Assuming that signal disposition is a per-thread attribute (big assumption 1) 
and assuming that signal disposition really works on all platforms (big 
assumption 2) we could ensure that the main (startup) thread got the signal if 
we had epicsThreadCreate call sigprocmask to block signals in all created 
(i.e. non-main) threads.  Of course dealing with signals that arrived right 
during the thread creation would probably still be an issue.

My recommendation to Bob Soliday remains, "try to find a way to deal with this 
situtation that doesn't involve catching signals".

--
Eric Norum                                 norume@aps.anl.gov
Advanced Photon Source                     Phone: (630) 252-4793
Argonne National Laboratory